Working Memory
Counting Span Task
Also called: CSPAN, counting span test, complex span task, dot counting span
Measure complex working memory with cards whose dark blue circles are counted while each card's total is held in mind and recalled in order.
12-15 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 25
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures working memory capacity with the counting span procedure. Participants count the dark blue circles on a series of cards, among squares and lighter circles, while remembering each card's total, then enter the totals in order.
Constructs
- Working memory capacity
- Executive attention
- Serial recall
- Counting under memory load
Research uses
- Individual-difference studies of working memory capacity
- Working memory research across ages and in groups who read little
- Within-study group comparison
How the paradigm works
The Counting Span Task measures working memory capacity: keeping numbers in mind while counting. On each card the participant counts the dark blue circles, among dark blue squares and light green circles, and chooses the number on a button from 1 to 9. After a set of 2 to 6 cards, the participant enters each card's number in the order of the cards, typed or tapped on a keypad.
More about the paradigm
The task is the counting span of Case, Kurland and Goldberg (1982), whose children counted green dots among yellow ones, in the form for adults of Engle and colleagues (1999) and Kane and colleagues (2004): 3 to 9 dark blue circles among shapes that share their color or their shape, so that finding a target needs both, counted aloud and then recalled in order, in sets of 2 to 6 cards. In Kane and colleagues' 236 adults its coefficient alpha was .77 (Conway and colleagues, 2005). Here the count is answered on buttons rather than aloud, and the procedure is the automated one of the reading and operation span tasks: 15 practice cards set each participant's time limit per card at the mean plus 2.5 SD of their counting times, three practice sets combine counting and recall, and the test gives three sets of each size from 2 to 6 in a random order, 60 cards. Counting accuracy is shown after each set, and accuracy below 85% is flagged. The scores use partial credit, as Conway and colleagues (2005) recommend: each number entered in its place counts. The cards are drawn by our code and have no words, so the test runs the same way in every language. No published norms are applied to this task. The scores support research and do not diagnose.
Scores
- Partial Span Scorecount
- Numbers entered in their place across all sets (0 to 60 at the standard settings).
- Higher is better
- Absolute Span Scorecount
- Sum of set sizes for sets recalled perfectly.
- Higher is better
- Partial-Credit Unit Score%
- Mean proportion of numbers entered in their place per set.
- Higher is better
- Counting Accuracy%
- Cards counted right, out of all cards shown; a card left past the time limit counts as wrong.
- Below 0.85 flags the session
- Counting Errorscount
- Speed errors plus wrong counts.
- Lower is better
- Speed Errorscount
- Cards not answered within the participant's time limit.
- Lower is better
- Wrong Countscount
- Cards answered with the wrong number.
- Lower is better
- Low Counting Accuracyflag
- 1 when counting accuracy is below 85%; review before analysis.
- Informational
- Recall, Sets of 2%
- Proportion of numbers entered in their place in sets of 2 cards.
- Higher is better
- Recall, Sets of 3%
- Proportion of numbers entered in their place in sets of 3 cards.
- Higher is better
- Recall, Sets of 4%
- Proportion of numbers entered in their place in sets of 4 cards.
- Higher is better
- Recall, Sets of 5%
- Proportion of numbers entered in their place in sets of 5 cards.
- Higher is better
- Recall, Sets of 6%
- Proportion of numbers entered in their place in sets of 6 cards.
- Higher is better
- Median Counting Timems
- Median time from a card to its answer, for cards answered within the time limit.
- Context for the time limit
- Counting Time Limitms
- The participant's time limit per card: mean practice counting time plus 2.5 SD.
- Informational
- Sets Completedcount
- Test sets recorded in the scored run.
- Fewer than 15 gets no percentile
Use and evidence
25 papers from 2021 to 2025 name this task in their title or abstract, up 19% compared with 2016 to 2020. That ranks it 382 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 6 |
| 2024 | 5 |
| 2023 | 2 |
| 2022 | 6 |
| 2021 | 6 |
| 2020 | 7 |
| 2019 | 5 |
| 2018 | 4 |
| 2017 | 4 |
| 2016 | 1 |
| 2015 | 6 |
| 2014 | 3 |
| 2013 | 4 |
| 2012 | 5 |
| 2011 | 2 |
| 2010 | 3 |
| 2009 | 3 |
| 2008 | 0 |
| 2007 | 2 |
| 2006 | 1 |
| 2005 | 2 |
| 2004 | 1 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 3 |
| 2000 | 0 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 1
- Papers on reliability or validity
- 8
- Registered clinical studies using it as an outcome
- 0
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Cards
- 3 to 9 dark blue circles, the targets, among 1 to 5 dark blue squares and 1 to 5 light green circles, one to a cell of a 6 by 6 grid, placed by the session's seed. Drawn by our code, with no words.
- Each card
- Count the dark blue circles and choose the number on a button from 1 to 9, before the time limit. A 500 ms blank follows.
- Time limit
- Each participant's own: the mean of their 15 practice counting times plus 2.5 SD, fixed for the session.
- Sets
- Three sets each of 2, 3, 4, 5 and 6 cards (standard; 1 to 3 of each size as a setting), 15 sets and 60 cards, in a seeded order. No two cards of a set have the same total.
- Recall
- After each set, the totals in the order of the cards, typed, or tapped on a keypad of 1 to 9 on touch screens. Untimed; every place needs a number.
- Feedback
- After each set, for 2,000 ms: the totals in their place, the counting errors in the set and the counting accuracy so far.
- Practice
- 15 cards counted alone, with feedback and no time limit, then three sets of 2 cards counted and recalled.
- Scores
- Totals in their place across all sets (partial-credit span), then sets recalled perfectly, counting accuracy and speed errors.
Case and colleagues (1982) and Kane and colleagues (2004) had people count aloud while an experimenter moved on to the next card. Here the count is answered on buttons, and each participant's practice times set a time limit per card, as in the automated reading and operation span tasks. No published norms are applied, and the scores support research and do not diagnose.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: test-phase answers to how many dark blue circles each card holds (nine buttons, chance 1 in 9; totals of 3 to 9 in about equal numbers) and speed errors, in the order shown, from the scored run. Counting accuracy below chance invalidates and not above chance is flagged. One button on 90% of answers invalidates when accuracy is not above chance. Speed errors are missed responses: more than max(3, 25%) invalidate, any is flagged. Answers under 200 ms on at least half the cards invalidate. Counting accuracy below 85% is flagged. Recall of the totals is not judged against chance. A hidden tab withholds the percentile.
Why these rules
Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). Typical participants make about one speed error on the automated span tasks (Redick et al. 2012), so more than a quarter signals non-engagement. Unsworth et al. (2005) and Redick et al. (2012) kept participants whose processing accuracy reached 85%, so accuracy below that is flagged. An answer under 200 ms is faster than a simple visual reaction, so it cannot reflect the item. Following response-time effort screening for untimed tests (Wise and Kong 2005), the session is invalid only when at least half the answers are that fast. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Sets of each size | 3 | 1 to 3 |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Untimed test: runs on desktops, laptops, tablets and phones. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Counting Span Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Case, Kurland & Goldberg, 1982; Kane et al., 2004), with the standard settings (sets of each size: 3). The main score was partial span score (numbers entered in their place across all sets (0 to 60 at the standard settings)).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Case, R., Kurland, D. M., & Goldberg, J. (1982). Operational efficiency and the growth of short-term memory span. Journal of Experimental Child Psychology, 33(3), 386-404. https://doi.org/10.1016/0022-0965(82)90054-6The counting span: count the targets on each card, then recall the totals in order.
- Kane, M. J., Hambrick, D. Z., Tuholski, S. W., Wilhelm, O., Payne, T. W., & Engle, R. W. (2004). The generality of working memory capacity: A latent-variable approach to verbal and visuospatial memory span and reasoning. Journal of Experimental Psychology: General, 133(2), 189-217. https://doi.org/10.1037/0096-3445.133.2.189The form for adults: dark blue circles among dark blue squares and light green circles, a 500 ms blank between displays, sets of two to six.
- Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm, O., & Engle, R. W. (2005). Working memory span tasks: A methodological review and user's guide. Psychonomic Bulletin & Review, 12(5), 769-786. https://doi.org/10.3758/BF03196772Partial-credit scoring, and the reliability of counting span.
- Unsworth, N., Heitz, R. P., Schrock, J. C., & Engle, R. W. (2005). An automated version of the operation span task. Behavior Research Methods, 37(3), 498-505. https://doi.org/10.3758/BF03192720The automated procedure: a time limit from practice and the 85% accuracy criterion.
Source papers
- Kane MJ, Hambrick DZ, Tuholski SW, Wilhelm O, Payne TW, Engle RW (2004). The generality of working memory capacity: A latent-variable approach to verbal and visuospatial memory span and reasoning. Journal of Experimental Psychology: General, 133(2):189-217.Paradigm source, Adults
Questions
How long does the Counting Span Task take?
About 12 to 15 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Partial Span Score, Absolute Span Score, Partial-Credit Unit Score, Counting Accuracy, Counting Errors, Speed Errors, Wrong Counts, Low Counting Accuracy, Recall, Sets of 2, Recall, Sets of 3, Recall, Sets of 4, Recall, Sets of 5, Recall, Sets of 6, Median Counting Time, Counting Time Limit and Sets Completed. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Counting Span Task used in research?
25 papers from 2021 to 2025 name it in their title or abstract, up 19% compared with 2016 to 2020. That ranks it 382 of 495 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.22.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Counting Span Task have?
In the published research, 1 paper reports its test-retest reliability and 0 registered clinical studies use it as an outcome (evidence grade D in the ConductCognition test census).
Which settings can a researcher change?
Sets of each size (1 to 3). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.
Related tests